The aim of this project is to assess the effects of aging at a behavioral level of analysis, to identify neurobiological mechanisms associated with these effects, and to evaluate interventions that might alter age-related performance decrements. Rodent and nonhuman primate models are tested in a battery of sensorimotor and learning/memory tasks. Neurochemical and neurohistological assays are conducted to determine neurobiological correlates of functional losses. Interventions include various nutritional, hormonal and pharmacological treatments. Multiple genotypes and genetically modified mice are examined to determine possible genetic involvement in age-related behavioral impairment. We have identified various effective pharmacologic strategies for improving learning performance of aged rodents using manipulations of the cholinergic and glutamatergic neurotransmitter systems. The lead cholinergic compound in this effort is the cholinesterase inhibitor, phenserine, which is currently in Phase III clinical trials using oral dosing in patients with Alzheimer?s disease (AD). Given the success of that effort, we have been evaluating a series of novel inhibitors of glycine uptake as well as D-serine, which binds to a strychnine insensitive glycine receptor, to overcome a maze learning impairment in rats induced by a pharmacological block of N-methyl-D-aspartate (NMDA) glutamate receptors. Inhibition of glycine uptake should enhance signaling through the NMDA receptor. Thusfar, we have not found an effective compound or combination of compounds. The objective of these studies is to design a single compound that can stimulate this receptor as well as inhibit cholinesterase. We believe that this strategy offers greater potential for cognitive enhancement than inhibition of either target alone. We have continued work examining the role of nitric oxide in memory dysfunction. Nitric oxide enhances guanylate cyclase and production of the second messenger, cyclic guanosine monophosphate (cGMP), which in turn stimulates glutamate release. Inhibitors of cGMP-specific phosphodiesterase enzyme type 5 (PDE-5) can enhance cGMP activity. Sildenafil citrate, the main ingredient in Viagra, is one such PDE-5 inhibitor in clinical use for erectile dysfunction. In addition, clinical reports of persons using this compound have indicated cognitive effects which have been demonstrated in animal learning experiments. We have begun to evaluate this compound on maze performance in rats. We observed that sildenafil can attenuate a maze learning impairment in young rats induced by antagonism of nitric oxide synthase, the enzyme involved in production of nitric oxide. We are currently investigating whether maze learning in aged rats can be enhanced by sildenafil treatment. We have also made progress this year in establishing a model of estrogen deprivation and replacement. Using the 14-unit T-maze, we found that young female rats were not impaired in learning following removal of their ovaries, but that replacement of estradiol did enhance retention of learning in this task. As a nutritional intervention, we have continued our investigations of the neuroprotective effects of a diet enriched with blueberries, which are high in polyphenolic compounds that can act as potent antioxidants. We found that rats maintained for 3 months on the blueberry enriched diet showed marked attenuation of the maze learning impairment induced by hippocampal injection of the neurotoxin, kainic acid (KA, which acts through glutamate receptors) compared to rats on a control diet, and that their improved performance was associated with reduced loss of hippocampal neurons. We also continue to assess the effects of a low calorie diet on various parameters of brain function as well as pharmaceuticals that mimic the effects of calorie restriction. As an example, we have found that rats on 40% calorie restriction (CR) show enhanced locomotor responses following injections of the dopamine agonist, amphetamine, compared to controls. Rats fed 2-deoxyglucose that mimics the effects of CR showed similar responses. We are currently completing similar studies that examine the effects of a high fiber diet (methylcellulose) and whether we observe similar CR mimetic effects when calories are reduced but the rats are presumably sated. Studies of neuromorphometric changes in mouse models of Alzheimer?s disease (AD) are also continuing including quantitation of changes in the length and complexity of capillaries and in particular populations of nerve fibers, specifically those containing tyrosine hydroxylase. Because mouse models of AD do not develop severe neurodegeneration, we have also been examining a new mouse mutant, known as the Harlequin mouse, that has a defective cell cycle that results in marked neuronal death in cerebellum and also in selected regions of the forebrain. We plan to do cross breeding of AD mouse mutants and the Harlequin mouse to determine if the AD genotypes will accelerate neurodegeneration. Regarding other progress for conducting neuromorphometrics using unbiased stereology, we are developing an automated method for counting neurons that are can be stained in high contrast to their background.